Navigating crisis: marketing dynamics and resilience in the MENA’s dual-banking system amidst the SAR-COV-2 pandemic

Abstract

In this paper, we investigate how market concentration and efficiency impact banks’ performance and stability during the SAR-COV-2 pandemic. There is a research gap in the empirical literature in understanding the specific impact of market concentration and efficiency on the profitability and risk-taking behavior of banks, particularly in the developing context of the Middle East and North Africa (MENA) region. To address this gap, we examine the relationship between market concentration, efficiency, and bank performance using a comprehensive dataset encompassing 575 banks across 20 MENA countries from 2006 to 2021. Specifically, we examine the market dynamics and performance of different banking systems co-existing in the MENA region. Our findings indicate that prior to the pandemic, both conventional and Islamic banks benefited from enhanced financial stability as a result of the increased market concentration and efficiency. However, during the pandemic, the positive effects of efficiency and concentration were primarily observed within the group of Islamic banks. Furthermore, we provide new evidence for the moderating effect of market concentration, with a significant negative impact suggesting that increased market competition reinforces the efficiency effect during the pandemic. Our findings are important for policymakers and regulatory authorities in the MENA region as they indicate the need for new policies that assign a more significant role to Islamic banking in the post-SAR-COV-2 recovery.

Introduction

This paper aims to augment the current body of literature on the financial performance and stability of banking systems in times of crisis. Our primary focus is on Islamic and conventional banking systems co-existing in the Middle Eastern and North African (MENA) region. Prior research (Xie et al. 2021; Miklaszewska et al. 2021; Rizwan et al. 2022) states that the current global (SAR-COV-2) crisis has had a direct impact on the real economy and led to several vulnerabilities and increased systemic risk in the banking sector, including the loss of credit due to non-performing loans, which posed a threat to banks’ corporate loan portfolios. Additionally, the pandemic negatively affected “institutions and individuals who may have a hard time liquidating their assets, these assets include limited access to credit” (Deloitte Insights, 2020, p. 2). As a result, the likelihood of default for financial institutions, including banks worldwide, has greatly increased. Therefore, studying the stability of the banking sector in the context of the current global (SAR-COV-2) crisis is of great importance to both scholars and policymakers.

Recent empirical literature offers some initial evidence supporting the significant impact of the SAR-COV-2 pandemic on the stability of banking systems across various countries and regions. Specific attention has been given to the stability and performance of financial institutions in a “dual-banking system” (i.e., Islamic and conventional). In this context, numerous studies have addressed the question of whether Islamic banking institutions are more resilient to the SAR-COV-2 pandemic compared to their conventional peers (see, e.g., Grassa et al. 2022; Fakhri and Darmawan, 2021; El-Chaarani et al. 2022, Rizwan et al. 2022). However, the results were mixed. For instance, Abdulla and Ebrahim (2022) reported “that GCC [Gulf Cooperative Council] banks were negatively affected by the pandemic. However, Islamic banks have performed better than conventional banks” (p. 239). According to the authors, this can be accredited to the Shariah principles that Islamic banking follows, which helped alleviate the pandemic’s adverse effects. In contrast, Grassa et al. (2022) concluded that “Islamic banks are not as profitable and resilient in the COVID-19 pandemic as in the global financial crisis (2007–2008)” (p. 251). Furthermore, according to the same research, Islamic banking institutions in the GCC countries have accumulated experience and demonstrated increased efficiency and stability over time. In a similar study, El-Chaarani et al. (2022) analyzed and compared the financial performance of Islamic banks (IBs) and conventional banks (CBs) in the GCC countries before and during the pandemic. The results of this study “reveal that there is a significant difference between Islamic banks and conventional banks during the crisis of COVID-19, where the conventional banks have presented a higher level of financial performance and financial liquidity than their Islamic counterparts” (El-Chaarani et al. 2022, p. 1). Hence, the evidence regarding the superior performance of IBs during the SAR-COV-2 pandemic remains inconclusive.

To fill this gap, our study investigates and compares the factors that determine the financial stability and performance of “different banking systems” in the MENA region. Particularly, we assess the impact of concentration and efficiency on bank performance before and during the pandemic. Since the analysis of the economic relationship between efficiency and profitability has received limited attention in the research literature, our study places special emphasis on examining efficiency as a determinant of bank profitability, rather than the other way around, highlighting its relevance in the context of banking system performance in the MENA region.

The theoretical framework that may explain the direction of the relation between “efficiency and profitability” is the so-called efficient-structure (ES) hypothesis (Demsetz, 1973). The ES hypothesis “entails the notion that the structure of the market may reflect differences in efficiency rather than a competitive situation” (Leon, 2015, p. 12). The hypothesis “also predicts that in concentrated market structure, the more dominant banks get the higher profitability with the increase of the efficiency” (Cristian et al. 2020, p. 408). In other words, as market concentration increases, the efficiency of the market also increases. Consequently, it is anticipated that more efficient banks will also be more profitable when operating in concentrated markets. Our expectations of a strong relationship between efficiency and profitability are supported not only by established economic principles but also by numerous relevant studies in this area. For instance, much of the empirical literature indicates that efficiency enables banks to enhance profitability, maintain financial stability, and mitigate the adverse effects of financial crises. Some of these studies utilize data from the US banking market (Assaf et al. 2019), China (Tan et al. 2017), and India (Rakshit and Bardhan, 2022), while others concentrate on developing regions with prevailing Islamic banking (Cristian et al. 2020; Mateev et al. 2022b; Mirzaei et al. 2024). Their findings confirm the positive relationship between efficiency and profitability. Conversely, several studies suggest that more efficient banks may be less profitable than their less efficient counterparts (see e.g., Kozak, 2021). Since the findings are ambiguous, more research on the efficiency-profitability paradigm is needed.

Another strand of empirical literature has investigated this relationship from a different perspective, such as employing profitability measures to predict efficiency. For example, a study by Otero et al. (2020) on the determinants of bank cost efficiency in the MENA region concluded that “cost efficiency is positively related to bank performance, but the level of concentration and market share has a negative influence on the former” (p. 1). Similar results were reported for developing economies in the “Southern, Eastern and Central Europe (SECE)” region (Kozak and Wierzbowska, 2021). However, our study takes a different approach by examining efficiency as a predictor of profitability, thus emphasizing its relevance in the context of a dual-banking system. The changes in banking operations, particularly in Islamic banking with respect to Shariah compliance, diversification, and business models, have made efficiency and competitive capability crucial factors in assessing bank performance, especially in the MENA region.

Since Islamic and conventional banks follow different business models and approaches to managing risk, the impact of the SAR-COV-2 pandemic is expected to be different. If this is the case, what could be the potential factors that may underpin the superior performance of either type of bank? Our analysis indicates that no prior empirical studies have examined the role of market concentration and efficiency in determining the financial performance and stability of banking systems in the MENA region. To fill this gap, this study explores the performance of banks in countries with a “dual-banking system”. More specifically, we identify and compare the impact of concentration and efficiency on bank stability and performance before and during the SAR-COV-2 pandemic. Our sample consists of 575 banks (both Islamic and conventional) located in 20 MENA countries, where the Islamic banking sector significantly commits to the economic advancement of these nations. The comparative analysis of the two banking systems reveals that, before the pandemic, both types of banks had benefited from enhanced financial stability as a result of increased market concentration and efficiency. However, during the pandemic, the positive effects of efficiency and concentration were only observed within the group of IBs.

Our research differentiates from previous studies conducted in this field in the following ways. First, prior research has primarily concentrated on the impact of SAR-COV-2 on the banking industry’s performance in specific countries and/or regions (Barua and Barua, 2021; Feyen et al. 2021; Colak and Öztekin, 2021; Miklaszewska et al. 2021; Rizwan et al. 2022). In contrast, our research’s main focus is on the banking systems co-existing in developing regions such as MENA and their comparative performance. Specifically, we explore the main factors such as concentration and efficiency, that influence bank performance and stability before and during the pandemic. Second, when assessing the stability of the banking industry, most studies tend to concentrate on the impact of either market structure or efficiency. Moreover, only individual analyses have been conducted on this matter (Rashid and Jabeen, 2016; Cristian et al. 2020; Mirzaei et al. 2024). In contrast, we analyze the combined effect of market concentration and efficiency on bank stability during the SAR-COV-2 pandemic considering it as an exogenous shock that significantly increased the risk of bank failure. Finally, the predominant method utilized in previous studies to assess bank performance typically involves accounting measures such as “return on assets, return on equity, and net interest margins” (Sun et al. 2017, p. 195). In addition to these ratios, we employ “aggregate stability and risk-adjusted performance measures” (Miklaszewska et al. 2021, p. 10), which significantly improves the reliability of financial stability analysis.

We believe the MENA region is an appealing laboratory for research for the following reasons. First, similar to other regions, the significance of Islamic finance and banking for the economic advancement of these regions, principally in countries with significant Muslim populations, has boldly increased. Islamic finance and banking became a significant part of the development agenda in the MENA region as well. Islamic banking provides multiple financial services, including “Qardh-Al-Hasan, Zakat, Waqf and Social Sukuk, for countering the adverse impact of COVID-19 on SMEs and individuals” (Syed et al. 2020, p. 11). Hence, it is reasonable to anticipate that the harmful impact of the SAR-COV-2 pandemic on Islamic institutions will be less severe compared to their conventional counterparts, primarily owing to their unique financial structures (Mirzaei et al. 2024; Viphindrartin et al. 2022). Moreover, IBs hold a dominant position in implementing the Basel III framework, thereby “ensuring sufficient risk protection through compliance with regulatory capital requirements, while CBs are striving to match the standards set by IBs” (Sun et al. 2017, p. 195). Therefore, investigating whether IBs in the MENA region are more resilient to the pandemic crisis vis-à-vis their conventional peers requires further research.

Second, regardless of the type of bank, the MENA region experiences “a monopolistic market where banks have significant market power”, as indicated by Mateev et al. (2022a). After the global financial crisis of 2007–2008, there was a significant increase in the “market power” of both types of banks, with IBs emerging as the dominant force in market power across the MENA region, as documented by Moudud-Ul-Huq et al. (2020). Since the increased market power is associated with more concentrated banking markets, the role of market concentration in determining the superior performance of IBs during the pandemic emerges as a crucial issue to explore. Finally, we acknowledge that “there are substantial differences in terms of financial, political, regulative, and economic features, among others, between developed and developing countries, and even within the MENA region” (Mateev et al. 2023, p. 2). Despite these differences, the outcomes of this study can be generalized to other regions and countries with the prevailing presence of Islamic finance. Moreover, the anticipated role of Islamic finance in the post-SAR-COV-2 period has become a crucial driver for economic recovery and growth in the MENA region post-pandemic (Hassan et al. 2020; Syed et al. 2020). Therefore, our findings may provide useful guidance for regulators and policymakers in other countries and regions worldwide, suggesting a more significant role of Islamic finance in the recovery from the SAR-COV-2 pandemic.

Our research makes a significant contribution to the banking field in several ways. Firstly, prior research has examined the financial performance of different types of banks in the MENA region in order to determine whether IBs exhibit “greater financial performance during the pandemic than their conventional counterparts”. Several papers also provided a comparative analysis of IBs and CBs regarding “significant differences in terms of systematic risk for conventional and Islamic banks before or during the SAR-COV-2” (Rizwan et al. 2022, p. 26).¹ We enhance the existing research by examining the main determinants of bank performance and stability in the competitive context of the MENA region, both before and during the pandemic. Our findings indicate that while in the pre-crisis period, market concentration and efficiency were important determining factors of bank performance for both types of banks, during the pandemic the positive effects of efficiency and concentration on bank stability were only observed within the group of IBs. One possible explanation is that these banking institutions possess a unique business model “in terms of their asset-liability structure and product offering” (Rizwan et al. 2022, p. 2) which renders them more resilient to the adverse effects of the SAR-COV-2 pandemic.

Secondly, we present new evidence for the moderating effect of market concentration, with a significant negative impact suggesting that increased market competition reinforces the efficiency effect during the pandemic. Our study also adds value to the ongoing discourse on the effectiveness of conventional business models. These models depend on intermediary services and income earned from interest. The debate centers around their continued performance and stability (Miklaszewska et al. 2021). Our study, which concentrates on bank performance and stability in the MENA region, provides new evidence in support of Li et al. (2021) findings, indicating that revenue diversification may enhance bank profitability during a crisis. Specifically, our analysis signifies that non-interest income has a negative effect on the capital ratios of MENA banks, suggesting that these banks do not rely heavily on non-interest revenue sources to mitigate the increased risks during the pandemic. Thus, the findings of this study call for adjustment of the current banking models that heavily rely on intermediation and interest-based earnings, particularly for conventional banks.

Thirdly, our research employs a methodology that is significantly different from the traditional approach of utilizing accounting- or market-data-based measures. We are the first study on the MENA region that employs “aggregate stability and risk-adjusted performance measures” to assess bank performance (Miklaszewska et al. 2021, p. 10). This approach significantly improves the reliability of financial stability analysis. Finally, our findings provide practical guidance for regulators and bank managers seeking to improve bank stability and performance. For instance, strategies that enhance efficiency and banking market concentration are especially crucial during turbulent times due to their positive effect on banks’ financial stability. These strategies are particularly relevant during the current financial crisis, marked by a strong upsurge in credit and liquidity risks as well as operational expenses. As Islamic banking institutions are more affected by these challenges, our results underscore the need to establish support policies aimed at facilitating their rapid recovery in the post-COVID-19 period.

The rest of the study is structured in the following way. Section “Literature review and formulation of the hypotheses” provides a summary of the findings from the most pertinent studies aligned with the goals of this research and presents the main hypotheses based on this analysis. Section “Data and methodology” describes the methodology, including the samples, variables, and model specifications. Section “Empirical analysis and results” illustrates the results and their interpretation. Section “Alternative tests and robustness checks” includes the robustness tests and their outcomes. Lastly, in Section “Discussions and conclusions”, we provide our conclusions, encompassing policy implications and addressing any research limitations.

Literature review and formulation of the hypotheses

Bank performance and stability in the MENA region: Islamic banking context

Since its introduction in the mid-1970s, the Islamic banking sector has acknowledged tremendous growth (Mallin et al. 2014). Currently, it possesses “71% (1.7 trillion $US) of the total Islamic financial assets” (Isnurhadi et al. 2021, p. 841) and is represented by 566 Islamic banks across 76 countries, with an additional 207 Islamic banking windows (Islamic Finance Development Report, 2022). The Islamic Finance Outlook (2022) edition emphasized that the “Islamic finance industry expanded rapidly in 2020 with total assets increasing 10.6% despite the double shock from the SAR-COV-2 pandemic and drop oil prices” (p. 4). On a positive note, S&P Global Ratings (2022) anticipates that “the global Islamic finance industry will grow by 10%–12% in 2021–2022” (p. 4).² Following the widespread adoption of the “new normal” by global economies, there has been a resurgence in economic activities due to the high demand for technological commodities, oil and gas, and the implementation of digitalization, particularly through the global Fintech transformation.³ This has also positively impacted the Islamic banking sector worldwide.

The recent surge in the popularity of Islamic finance can be attributed to various factors. Yilmaz and Gunes (2015) explain that the continuous interest of policymakers and regulators worldwide has contributed to the rapid spread of Islamic banking. Meanwhile, Bitar et al. (2020) point out that the sector’s instant development is largely driven by two factors: the aspiration amongst the Muslim population to spread the Sharia law across all economic endeavors and the oil revenues from Gulf countries. Sharia law guides all transactions in Islamic banking, which operates in highly regulated environments (Shawtari et al. 2019). As interest is forbidden in Sharia law, IBs rely on transactional and intermediation contracts to earn their profits (El-Hawary et al. 2004). Typical types of transactions, or use of funds, include “leasing, purchase and resale transactions (Murabaha(h) and Ijarah) or profit and loss sharing (Mudarabah, trust financing or limited partnership and Musharakah, joint venture, where investment are not required to be paid back)” (Brown et al. 2007, p. 2). Moreover, “the relationships between IBs and their customers are based on mutual trust, strengthened by shared religious beliefs” (Brown et al. 2007, p. 2). A key aspect of the social significance of IBs compared to their conventional peers is the ethical considerations. In other words, beyond their distinctive business model, IBs are expected to have a stronger social standing due to their adherence to ethical principles.

However, academic literature lacks consensus on the performance and financial stability of IBs. Some researchers argue that “Islamic banks are, on average, more profitable, more liquid, better capitalized, and have lower credit risk than conventional banks” (Ben Khediri et al. 2015, p. 75), and therefore, are more stable (Beck et al. 2013; Abedifar et al. 2013). Other researchers, (see, e.g., Kabir Md. and Worthington, 2017 among others) claim that IBs are less stable than CBs in times of crisis. For instance, Abu-Alkheil et al. (2017) reported “that CBs grant a higher percentage of loans relative to their peers and are more profitable than IBs in the pre-, during-, and post-crisis periods. Additionally, in the post-crisis period, IBs held more liquidity and experienced a slower recovery. However, their profitability gap with CBs is narrowing” (p. 1). In general, previous studies on IBs performance and stability have yielded inconclusive results regarding their superior performance. Therefore, further research is needed to better address this issue.

When comparing IBs to their traditional counterparts, various factors may account for the differences in their performance. Numerous studies have shown that the primary drivers of bank performance are market concentration and operating efficiency. For instance, Rashid and Jabeen (2016) conducted an empirical study in Pakistan that “examined the bank-specific, financial, and macroeconomic determinants of bank performance in both Islamic and traditional banks” (p. 1). The study revealed that “operating efficiency, reserves, and overheads are significant determinants of conventional banks’ performance, whereas operating efficiency, deposits, and market concentration are significant in explaining the performance of Islamic banks” (Rashid and Jabeen, 2016, p. 92). From a theoretical point of view, the “efficiency-structure” (ES) hypothesis provides insight into the link between “market concentration” and “profitability”. This concept argues that the more efficient banks gain higher market share and profitability, thus leading to a concentrated market. Thus, banks operating in concentrated markets are anticipated to be more profitable.

Regarding the efficiency effect, González et al. (2017) claimed that “in less-competitive markets, increased competition may favor the risk-shifting effect and help improve efficiency, which in turn improves financial stability” (p. 592). Other studies also suggest that enhanced efficiency positively influences the stability of a bank, but “this effect depends on the level of market competition” (Saeed and Izzeldin, 2016; Mateev et al. 2022b; Isnurhadi et al. 2021). However, no empirical research has so far analyzed the differential effect of concentration and efficiency on banks in a “dual-banking system”, where banks compete “based on different business models and risk management practice” (Mateev et al. 2022b, p. 12). Our study contributed to this literature by investigating the main factors that explain the discrepancies in bank performance between Islamic vs conventional banking in the developing context of the MENA region.

Islamic banking stability and performance during SAR-COV-2

As noted by previous studies, CBs and IBs share similarities in their banking practices, but their objectives and operational nature differ significantly, including the pricing of their products. Sun et al. (2017) contends that “while CBs rely on market-determined interest rates to establish lending and borrowing rates, IBs are believed to base this decision on future expected profit-sharing yields rather than market-based interest yields” (p. 195). The distinction between the two banking systems is rooted in the fact that CBs are compelled to adhere to local rules, regulations, and Basel III requirements, while IBs are “required to comply with Shariah principles in both the design of products and also in continually monitoring compliance to these rules from this additional supervisory layer” (Sun et al. 2017, p. 195).

Therefore, extensive research has been conducted to understand why IBs have shown greater resilience to the COVID-19 pandemic compared to their conventional peers. According to studies conducted by Mirzaei et al. (2024) and Viphindrartin et al. (2022), IBs’ unique financial structures may be the cause of their resilience. For instance, IBs’ funding bases are more diversified than those of CBs, and they are less reliant on short-term wholesale funding. IBs also tend to have lower levels of leverage, making them less vulnerable to market fluctuations. The principles of Islamic finance prioritizing risk-sharing and asset backing also aid in mitigating the negative impact of the economic downturn on IBs’ loan portfolios. Nonetheless, this observation is not applicable across the board since the performance of IBs varies depending on the country and individual bank’s approaches to the SAR-COV-2 pandemic (El-Chaarani et al. 2022). Supporting this argument, Mansour et al. (2022) examined the impact of the SAR-COV-2 pandemic on IBs using aggregated assessment of “size, profitability, nonperforming financing, and stability” (p. 265), and reported that the impact of SAR-COV-2 differs across countries in their sample. The study concluded that the measures to be taken by policymakers should not be standardized but instead prioritized based on the IBs short-term response to the pandemic, the country’s economic condition, and macroeconomic aspirations.

Several factors have been identified as contributing to the superior performance of IBs during the SAR-COV-2 pandemic, including Shariah compliance, diversification, and business models. For instance, Alabbad and Schertler (2022) examined the income and banks’ stock prices responses to “the COVID-19 policy measures in countries with the dual-banking system” (p. 1511) and reported “that the Shariah compliance does not limit the adverse impact of the COVID-19 crisis on Islamic banking, but that IBs performance responds more positively to income support initiatives than the one of conventional banks” (p. 1). Le et al. (2022) investigated “the relationship between diversification and Islamic banking systems’ performance under the impact of the COVID-19 turmoil” (p. 1). The study confirmed the adverse effects of the SAR-COV-2 shock and reported that “income diversification is found to mitigate the adverse effect of this health crisis on the performance of the Islamic banking systems” (Le et al. 2022, p. 1). Finally, Boubaker et al. (2023) examine “whether the Islamic banking business model makes corporate earnings more uncertain [..] with 15 years of data for 532 banks (129 Islamic and 403 conventional) from 23 Muslim countries across the world” (p. 1). The findings suggest that due to higher operational costs, IBs’ return on assets is more uncertain than that of CBs. Furthermore, the research supports previous evidence indicating that “Islamic banks generally have fewer nonperforming loans than conventional banks” (p. 1).⁴

Upon scrutinizing the literature on the topic, several key findings have emerged. Firstly, the majority of earlier studies demonstrated that the SAR-COV-2 pandemic had a significant and unfavorable impact on the performance of both Islamic and conventional banking. However, the prior research produced inconsistent results regarding IBs’ superior performance during the pandemic. Secondly, scholars used various factors, including Shariah compliance, diversification, and business models, to explain the differences in the performance of CBs and IBs during the pandemic crisis. Yet, the significance of market concentration and efficiency in maintaining the stability of IBs during the recent financial crisis has not yet been fully explored. To bridge this gap, we analyzed both the individual and combined effects of concentration and efficiency on banks’ performance and stability in the MENA region. Furthermore, in the next section, we tested several hypotheses regarding the significance of these effects assuming they are more pronounced for IBs and during the SAR-COV-2 pandemic.

Development of hypotheses

Undeniably, the SAR-COV-2 pandemic and the resulting financial crisis have significantly impacted the way banks generate profits and assess risks, resulting in a shift in their business models (Carletti et al. 2020). Studies that investigate bank performance have identified several factors that determine the stability of banks, with a particular focus on competition and efficiency in the banking market. These findings align with either the “efficient-structure” (ES) or the “structure-conduct-performance” (SCP) hypotheses. The ES hypothesis posits that firms which are efficient have the ability “to generate higher profits, leading to an increase in size and market share and resulting in higher market concentration”. This, in turn, can be “explained by lower costs achieved through either superior management or production processes” (Goldberg and Rai, 1996, p. 746). Therefore, efficiency leads to a concentrated market. In contrast, the SCP theory postulates that “high market concentration and market power motivate banks to set favorable prices and achieve higher and extraordinary income” (Kozak and Wierzbowska, 2021, p. 40). Empirical studies tested the SCP hypothesis by “examining the relationship between profitability and market concentration with a positive relationship indicating non-competitive behavior in concentrated markets” (Goldberg and Rai, 1996, p. 746). In line with the SCP theory, we can assume that market concentration will positively affect bank performance. However, it remains unclear whether these relationships depend on the type of the business model (i.e., Islamic vs. conventional). Most previous studies suggest that IBs banks are less vulnerable to the SAR-COV-2 pandemic compared to their conventional peers (Alabbad and Schertler 2022; Aliani et al. 2022). Therefore, we expect that the superior performance of IBs during the pandemic is influenced by “the level of market concentration and efficiency” in these markets.⁵ In order to assess this proposition, we test the following hypotheses:

H1a: Market concentration and efficiency exert a significant influence on MENA banks’ profitability.

H1b: This effect is anticipated to be stronger for IBs and during SAR-COV-2.

While prior research has explored the topics of efficiency and financial stability separately, there are few studies examining the paradigm of efficiency-default risk. For instance, Saeed and Izzeldin (2016) conducted “a comparative analysis of conventional and Islamic banks in the GCC countries, discovering that the absence of a trade-off between efficiency and stability for IBs suggests that efficiency and default risk are plausible early warning indicators of IBs instability” (p. 1). However, it is possible that higher efficiency levels could lead to elevated risk levels, ultimately exposing banks to financial difficulties. Moreover, the SCP theory posits that “market concentration or market power is the primary factor driving a bank’s superior performance” (Goldberg and Rai, 1996, p. 746). As a result, banks operating in concentrated markets are less likely to collide in order to achieve higher profits. Consequently, these banks are willing to accept more risk. The empirical literature provides inconclusive evidence of these effects. While Miklaszewska et al. (2021) reported “a positive but insignificant relationship between market concentration and a bank’s probability of default” (p. 1), Mateev et al. (2022a) claimed a “positive and nonlinear association between market concentration and bank stability for CBs, but a negative for IBs” (p. 10). However, there is no evidence supporting the “efficiency-default risk” paradigm in a comparative setting that includes IBs. Whether this differential effect is more pronounced during the recent financial crisis remains unclear. Therefore, to address this puzzling issue, we test the following hypotheses:

H2a: MENA banks’ stability is positively associated with marker concentration and efficiency.

H2b: This effect is anticipated to be stronger for IBs and during SAR-COV-2.

The banking industry’s conduct has undergone a considerable shift in public perception following “the global financial crisis of 2007–2008.” This event has highlighted the significance of maintaining banking system stability and controlling risk (Allen et al. 2009), as well as ensuring an adequate level of capitalization (Demirgüç-Kunt and Huizinga, 2010). An increase in the capital ratios of banks has been observed in various regions and countries after the crisis, which is largely due to “increases in capital, the issuance of new equity and/or retained earnings, and assets declining” (Carletti et al. 2020, p.48). Prior research has explored the paradigm of capital-to-bank stability and attempted to evaluate the impact of efficiency on bank risk. For instance, Isnurhadi et al. (2021) studied “the relationship between bank capital, efficiency, and risk in Islamic banks” (p. 841). The research reported that “the interaction test of bank capital and efficiency shows that efficiency encourages banks to reduce risk, including when bank capital is relatively lower” (Isnurhadi et al. 2021, p. 841). However, the association between market concentration and bank capitalization remains ambiguous. In their study on the “Central, Eastern, and Northern European Countries (CENE)” region, Miklaszewska et al. (2021) found “no evidence for a significant correlation between market concentration and bank capitalization” (p. 18). However, their analysis indicates that “bank size, loan dynamics, and non-performing loans portfolios are negatively correlated with banks’ capital ratios” (p. 17). In the context of the MENA region, the anticipated consolidation prompted by the pandemic may diminish competition and foster the creation of oligopolistic market structures. (Hamadi and Awdeh, 2020). As a result, SAR-COV-2 may exacerbate the impact of market concentration on banks’ capital ratios, thereby enhancing their financial stability. To test this proposition, we frame out the last two hypotheses in the following way:

H3a: Market concentration and efficiency exhibit a significant correlation with banks’ capital ratios.

H3b: This effect is anticipated. to be stronger for IBs and during SAR-COV-2.

Data and methodology

Sample selection

Our data set spans the period from 2006 to 2021 and contains banks from 20 MENA countries, including the GCC countries. Our main source of bank accounting data is the Orbis BankFocus (Bureau van Dijk) database. Furthermore, we compiled data from annual reports available on bank websites to establish a comprehensive database for banking institutions in the MENA region. We also utilized the World Bank database, specifically the “World Development Indicators”, to gather macroeconomic data. To ensure that our analysis covered both pre- and SAR-COV-2 pandemic periods and that we had sufficient yearly data availability, we selected a specific time frame. This allowed us to gain a deeper understanding of how market concentration and efficiency affected bank performance and stability. We excluded banks with missing accounting information, resulting in a data set of 575 banks (both Islamic and conventional). One of the main benefits of our sample is its diversity in terms of country, bank type, and income group (see Table 1). Our data set is unbalanced and includes a larger number of conventional (471) than Islamic (104) banks. Following previous studies, “all the variables are winsorized at the 1% and 99% levels to mitigate the effect of outliers” (Bitar et al, 2016, p. 11).

Table 1 Composition of banks by country.

Econometric model

To assess the association between market concentration, efficiency, and bank performance (profitability and risk), we employ the following estimation model:

$$\begin{array}{l}{{BP}}_{it}={\vartheta }_{0}+{\beta }_{1}{\rm{Concentration}}_{it}+{\beta }_{2}{\rm{Efficiency}}_{it}\\\qquad\quad+\,{\beta }_{3}{\rm{Concentration}}\,*\, {\rm{Efficiency}}_{it}\\ \qquad\quad\times \,{\beta }_{4}GC{C}_{t}+\psi {X}_{it-1}+{\upsilon }_{it}\end{array}$$

(1)

$$\begin{array}{l}{{BC}}_{it}={\vartheta}_{0}+{\beta }_{1}{\rm{Concentration}}_{it}+{\beta }_{2}{\rm{Efficiency}}_{it}\\\qquad\quad+\,{\beta }_{3}{\rm{Concentration}}\,*\, {\rm{Efficiency}}_{it}\\ \qquad\quad\times \,{\beta }_{4}GC{C}_{t}+\psi {X}_{it-1}+{\upsilon }_{it}\end{array}$$

(2)

In this setting, BP_it is the measure of the financial performance (profitability and risk) of bank i in year t; similarly, BC_it is the measure of the capitalization level of bank i in year t. We use “three accounting measures” of bank profitability (return on assets, ROA, return on equity, ROE, and cost-to-income ratio, C/I). In addition to the traditional bank stability measure (Z-score), we employ two comprehensive measures of bank performance and stability. Specifically, we compute the Multilevel Performance Score (MLPS) index and the Financial Stability Indicator (FSI), following the methodology proposed by Miklaszewska et al. (2021). The bank’s capitalization level is proxied by the capital adequacy ratio (CAR), equity to total assets (E/TA) ratio, and total capital requirements (TCR). Guided by Mateev et al. (2022b) “banking market concentration is proxied by the Herfindahl-Hirschman index (HH-index) where the index is a measure of market concentration calculated as the sum of the squared market shares for each bank in a country” (p.15). As an alternative measure of concentration, we employ “the share of top five banks in total assets of a country’s banking sector (CR5). X_it-1 is a vector comprising “bank-specific and country-specific variables” that are widely acknowledged in studies focusing on empirical research in banking as “significant determinants of bank performance”. The GCC dummy variable indicates whether a bank belongs to any of the GCC countries. It equals 1 if the bank is a member and, 0 otherwise. In our robustness tests, we employ a crisis dummy variable indicative of a systematic crisis, in our case, the SAR-COV-2 pandemic. Table S1 (see Supplementary Materials) explains the meaning of all the variables employed in the model and their data sources.

Following prior research (Abedifar et al. 2013; González et al. 2017; Mateev et al. 2022a; Mateev et al. 2023), we utilized fixed effect/random effect specifications in our analysis. We performed a Hausman test to make our preferred model choice between random effects and alternative fixed effects. Following Mateev et al. (2022a), “the choice between random and fixed effects specification depends on the Prob > chi² being more or less than 5% (if Prob > chi² is <0.05 use fixed effects)” (p. 10). To address the issue of time-related factors that may influence bank performance, we included “year dummy variables” in our reference model and conducted a robust Hausman test. The year dummies show significance at usual levels, suggesting temporal variations during the observation period.⁶ For robustness purposes, we employed the “Generalized Method of Moments (GMM)” system estimator, which is known to produce more precise and durable outcomes than FE/RE specifications (García-Herrero et al. 2009). GMM estimator is commonly used in financial research, specifically in studies regarding banking. Methods based on the GMM estimator “are particularly useful for models incorporating endogenous or predetermined explanatory variables” (Miklaszewska et al. 2021, p. 12).

Measures of bank profitability and risk

This study explores various indicators of bank performance (profitability and risk), which are widely utilized in the empirical literature. We employ “the return on asset ratio (ROA), which reflects how effectively banks generate profits from the total assets, and the return on equity ratio (ROE), which assesses the return on the shareholder’s equity, both as a proxy of bank profitability” (Mateev et al, 2023, p. 15). The “cost to income ratio” (C/I) is included “to control for any cross-bank differences in terms of efficiency; a higher value indicates a lower level of efficiency” (Bitar et al. 2016, p. 5).

According to prior research, it is relevant to assess the efficacy of banking institutions not only through individual ratios, but also through “aggregate financial stability indices, considering different aspects of banks’ financial strength such as profitability and capital adequacy, and major risks - credit and liquidity risk” (Kocisova, 2015, p. 1). Therefore, in addition to the conventional financial stability indicator (distance to default), we utilize two comprehensive measures of bank stability and risk-adjusted performance: “Multi-level Performance Score” (MLPS) and “Financial Stability Indicator” (FSI). The “distance to default or Z-score”, measures the proximity to insolvency (Roy, 1952), and “is determined by combining the bank’s Return on Assets (ROA) with the capital-to-asset ratio, and then dividing it by the standard deviation of ROA over a five-year period” (Issa et al. 2024, p. 15). It is well known that a higher Z-score signifies reduced exposure to insolvency. Several methods exist for constructing the Z-score measure, which are elaborated in detail by Lepetit and Strobel (2013). For this particular study, we follow Beltratti and Stulz (2012) and compute the Z-score over a period of 5 years ([t − 4, t]) on the return on assets (ROA) and its standard deviation (σROA) via a rolling window. Because of the skewed distribution of this measure, a natural logarithm transformation is implemented.

The computed performance measures for MENA banks during the 2006–2019 period, as well as the 2020–2021 figures, are reported in Supplementary Materials. Specifically, Table S2 summarizes the results for Z-Score, FSI, and MLP measures and Table S3—the results for profitability and equity ratios for the same two periods. From the analysis of the data in these two tables, we can conclude that the SAR-COV-2 pandemic has mostly affected bank profitability, whereas financial stability and bank capitalization remain relatively strong. A noteworthy improvement in the aggregate bank performance has been observed between 2013 and 2019, followed by a significant drop in 2020 and 2021, as illustrated in Fig. 1. All the performance measures used in this analysis (MLP-score, FSI, and LogZ) are weighted by total assets.

Fig. 1

Dynamics of bank stability and aggregate performance indicators during the period 2006–2021.

Measures of efficiency and market concentration

To calculate the efficiency scores, we employed “Data Envelopment Analysis (DEA)”, which was originally conceived by Charnes et al. (1978). According to Mirzaei et al. (2024), “DEA is a linear programming-based method of comparison that involves benchmarking multiple decision-making units (DMUs) with multiple inputs and outputs. DEA estimates the production possibility frontier and evaluates the efficiency of each DMU against the frontier” (p. 338). In this approach, “the analysis determines the optimal efficiency frontier by identifying the most efficient DMUs that achieve the highest level of outputs while minimizing inputs. The efficiency scores obtained from DEA range from 0 to 1, where a score of 1 indicates that a DMU operates at maximum efficiency” (Issa et al. 2024, p. 12). To calculate “DEA efficiency scores”, we followed the intermediation approach proposed by Chaffai and Hassan (2019). Table S4 (see Supplementary Materials) details ‘the intermediation approach for the inputs and outputs” used to calculate the efficiency scores. Following Issa et al. (2024), we utilize “three proxies of bank efficiency using input-oriented and output-oriented DEA models” (p. 12), namely, constant returns to scale (CRS), variable returns to scale (VRS), and SCALE (which is computed as the ratio of the first two). According to the empirical literature, “a score greater than 1 indicates that the bank is operating more efficiently than the average bank in the sample, while a score less than 1 indicates lower efficiency (Issa et al. 2024, p. 12). Consistent with previous research, we anticipate a positive correlation between efficiency and bank performance.

The concentration level in the banking market is assessed using the “Herfindahl-Hirschman index” (HHI). According to Hamza and Kachtouli (2014), this measure “is calculated by summing the squares of the market shares of every bank in the market or a country, and it varies between zero (situation of pure and perfect competition) and 10,000 (100²: monopoly position)” (p. 35). Table S4 (see Supplementary Materials) details index definition and the estimation approach. Our second concentration measure is “the share of top five banks in total assets of a country’s banking sector (CR5).” Previous research has shown that market concentration can impact bank performance, either positively or negatively. For instance, a study by Tan et al. (2017) utilized “the Lerner index and a three-bank concentration ratio to test further the impact of competition on bank profitability” (p. 2) in the Chinese banking sector. The study reported that “the concentration ratio is significant and negative indicating lower concentration leads to higher bank profitability which is different from the results reported from Lerner index” (Tan et al. 2017, p. 13). However, Agustini and Viverita (2012) asserted that “concentration has a positive effect on bank performance as postulated by structure-conduct performance (SCP) hypothesis” (p. 39). Finally, González et al. (2017) reported that “a positive relation between concentration and Z-score is observed in non-Gulf countries, and the opposite for Gulf countries” (p. 601). Therefore, the sign of this relationship remains unclear.

Control variables

In our analysis, we utilize key characteristics of banks that have been identified by Miklaszewska et al. (2021), Mateev et al. (2022b), and Mirzaei et al. (2024) as important factors in determining bank performance and stability. These factors include loans, non-interest income, non-financial loan growth rate, liquid assets, size, capital adequacy ratio, non-performing loans, and leverage (equity to total assets). As control variables, we subject these characteristics to our regression analysis. Previous research has suggested that size is a crucial determinant of bank performance, but its precise impact on profitability is uncertain. For instance, Blatter and Fuster (2022) found evidence supporting the economies of scale effect, as they discovered that efficiency and profitability tend to increase with bank size for most Swiss banks in their sample. However, other studies suggested that “large banks have lower cost efficiency, which indicates the presence of diseconomies of scale for banks in OPT” (Sarsour and Daoud, 2015, p. 55). The factor “contributing to diseconomies of scale is that larger banks become difficult to manage” (Yeddou, 2023, p. 10). In overall, the analysis of previous studies suggests that larger banks should exhibit better performance. Therefore, we assume a positive correlation with bank performance.

To estimate a bank’s financial strategy, we employed “the ratio of net loans to customer deposits” (Kosmidou, 2008, p. 1). This measure presents “the relationship between comparatively illiquid assets (e.g., loans) and comparatively stable funding sources (e.g., deposits and other short-term fundings)” (Kosmidou, 2008, p. 1). A low percentage indicates that the bank has sufficient liquidity to manage anticipated outflows related to loans. As a result, we predict that liquidity risk will have a positive association with bank performance. To control the bank’s level of income diversification, we employ “non-interest income to total assets ratio”. In accordance with prior research findings (Nguyen, 2012; Miklaszewska et al. 2021), we anticipate that bank stability and performance will exhibit a positive correlation with this variable. Following Neto et al. (2021), we measure banks’ risk preferences by using two ratios, “the leverage ratio (equity to total assets) and the liquidity ratio (liquid assets to total assets)” (p. 105). The liquidity ratio serves as an indicator of a bank’s ability to manage financing difficulties and reduce its balance sheet, as highlighted by Beltratti and Stulz (2012). It is our prediction that there should be a positive correlation between a bank’s liquidity and its overall performance.

Additionally, our analysis includes three widely used macroeconomic indicators, “the annual percentage growth rate of GDP, the inflation rate, and the percentage of domestic credit to the private sector as a percentage of GDP”, with the aim of accounting for differences in banks’ external environments. Previous studies have shown that economic growth is positively associated with bank profitability, suggesting that higher GDP growth will likely lead to improved bank performance. Conversely, inflation tends to increase bad debts, forcing banks to incur extra expenses to manage credit risk, which ultimately reduces their efficiency level (Pasiouras, 2008). This may adversely affect the bank’s profitability. However, Bourke (1989) and Molyneux and Thornton (1992) reported a positive association between inflation and bank profitability. Therefore, the net impact of the inflationary index remains ambiguous. We employ domestic credit as a percentage of the GDP ratio as a banking sector development indicator and assume a positive association with bank profitability (Miklaszewska et al. 2021).

Empirical analysis and results

Descriptive summary

Table 2 shows the “descriptive statistics” for the banks in our sample. The ratios were computed and compared across different banking systems in the MENA region (i.e., Islamic vs. conventional). Based on the results presented in the table, we derive valuable conclusions regarding the performance of each banking system.

Table 2 Descriptive statistics of banks.

First, we observe that CBs are more profitable over the entire observation period (2006–2021); both ROA and ROE exceed the profitability levels of IBs. In terms of aggregate performance, the higher value of the composite measure MLPS shows a better assessment of bank performance for IBs (1.69 vs. 0.83). However, the short-term stability indicator (FSI) takes a negative value for IBs (a mean of −0.02) but a positive value for CBs. To better understand the reasons for these results, we calculated the FSI for each country in the sample and found that the FSI values were not homogeneous across the sample (see Table S2). For some countries, the values are higher in the 2020–2021 period than in the preceding years (e.g., the Syrian Arab Republic and Yemen), indicating a strong and stable financial position of banks in these countries. For others (e.g., Iran and Tunisia), we found evidence of substantial credit risk related to the non-performing loan (NPL) portfolio (both before and during the pandemic) that translates into negative values of the FSI index. The measure of efficiency (DEA score) displays comparable values for both CBs and IBs, with a mean difference significant at the 1% level. Regarding financial stability, our analysis indicates that CBs are riskier than IBs, as demonstrated by the “distance-to-default, or Z-score measure” (2.51 vs. 2.84). Our further analysis of banks’ riskiness prior to and during the SAR-COV-2 pandemic shows that the primary reason for this outcome is the superior financial stability of IBs during the pandemic period (2020–2021), as evidenced by a higher Z-score (3.13 compared to 2.80).

Drawing from the data provided in Table 2, it can be noticed that the capital adequacy ratio (CAR) ranges from 21% to 25% (on average) across the sample banks. This number exceeds the minimum requirement for capitalization indicated by Basel agreements. Nevertheless, IBs appear to be better capitalized than CBs, with 25.5% compared to 21.5%. These findings remain consistent when comparing the total capital requirements (TCR) across different types of banks. Upon analysis, it was found that IBs exhibit a higher loan-to-total asset ratio and outperform CBs in terms of loans as a percentage of total deposits (66.3% vs. 60.9%). However, it is difficult to determine whether IBs performed better as the mean difference between the two groups is statistically insignificant. On the other hand, CBs hold more liquid assets (44.2% vs. 34.9%), while the level of bank leverage, assessed by the “equity-to-total assets ratio,” appears to be comparable in both samples. The non-financial loan growth rate, which reflects changes in the supply and demand of loans to non-financial institutions, is negative in both samples. Additionally, we provide estimations for the level of concentration of the MENA banking market, using the Herfindahl-Hirschman Index (HHI) and the concentration ratio (CR5) as proxies. The average HHI of the two samples is 0.20, indicating moderate concentration, with a statistically significant mean difference at the 1% level. The level of concentration measured by the CR5 ratio is relatively high for both IBs and CBs (around 60%).

The effects of market concentration and efficiency on bank profitability

The following section discusses the outcomes of bank performance evaluations conducted with the model specifications outlined in Eq. (1). Bank profitability served as the dependent variable, and two distinct metrics were used to determine profitability (ROA and ROE), alongside the cost-to-income (C/I) ratio, which was utilized as a measure of cost-effectiveness.

Tables 3 and 4 showcase the findings for various banking systems (IBs and CBs) and their performance during the pre-pandemic and pandemic periods, respectively. Specifically, data in Table 3 for the pre-pandemic period indicate that IBs profitability (measured by ROA) is strongly determined by both market concentration and efficiency (see Model 4) while these effects are insignificant within the sample of CBs. The same effect is observed for ROE, as reported in Models 2 and 5. Therefore, we find partial support for our first hypothesis (H1a). However, the effect of concentration and efficiency on cost-effectiveness is significantly negative (see Models 3 and 6). This suggests that concentrated banks and those with better efficiency are able to improve their level of operating effectiveness, regardless of their business model.

Table 3 Panel regressions of bank profitability (CB vs IB, 2006–2019).

Table 4 Panel regressions of bank profitability (CB vs IB, 2020–2021).

Our findings for the limited role of concentration in explaining the CBs profitability align with those reported for other developing regions (see, e.g., Miklaszewska et al. (2021) for the CENE region). However, the profitability of IBs is positively associated with efficiency. This finding is consistent with earlier studies that highlight efficiency as a critical factor influencing bank performance (Cristian et al. 2020; Mateev et al. 2022b). As a result, it appears that, before the pandemic, banks in concentrated markets and those with higher efficiency were able to achieve better performance. Since these effects are significant only for IBs, we confirm the validity of the H1b hypothesis. Next, we explore the “moderating effect” of market concentration on the relationship between efficiency and bank performance. The interaction term’s coefficient is significant and negative only within the sample of IBs (refer to Model 4). This suggests that in concentrated markets, the positive impact of efficiency on bank performance weakens. The reason is that concentrated markets are typically less competitive when banks’ market power is high. According to the SCP paradigm, market concentration encourages collusion between firms. Therefore, firms that operate in concentrated markets (characterized by a few dominant firms) may generate more substantial profits, regardless of their efficiency (Molyneux and Forbes, 1995). In contrast, our findings for the MENA region demonstrate that banks may benefit from their improved efficiency if they operate in less concentrated markets (banks with lower market power). The predominant impact of concentration and efficiency on IBs suggests that policymakers may consider customized measures, focusing on the IB’s prompt response to the pandemic, the economic situation of the country, and any government support initiatives, rather than standardized solutions.⁷

When examining the control variables, we can see a noteworthy impact of several bank characteristics. These include loans, non-interest income, size, non-financial loan growth, liquid assets, and non-performing loans. Our finding suggests that key bank characteristics determine the profitability of MENA banks, regardless of their business models. Liquidity, as measured by loans-to-deposits ratio, demonstrates a positive correlation with bank profitability (ROA and ROE), while negatively influencing a bank’s cost efficiency. Non-interest income and bank size positively affect both types of banks, while non-financial loan growth and liquid assets are significant only within the sample of CBs. Bank size has a notably positive effect on profitability, as identified in previous research by Flamini et al. (2009), who argue that larger banks have the potential to gain efficiency advantages and earn higher revenues due to operating in less competitive markets. As expected, an increase in non-performing loans decreases profitability, but it also leads to an increase in banks’ cost inefficiency. Regarding the impact of our “macroeconomic variables”, the business cycle (GDP growth) negatively affects the profitability of both types of banks, while inflation and domestic credit to the private sector have no significant impact. These results corroborate the findings of some previous studies (Mateev et al. 2022a, Miklaszewska et al. 2021).

Table 4 illustrates the results of bank performance during the SAR-COV-2 pandemic. Our analysis reveals that market concentration and efficiency are crucial factors influencing the profitability of CBs in the MENA region during the pandemic; however, this effect is insignificant for IBs. This finding contradicts the first hypothesis (H1b), suggesting that market concentration and efficiency would have a stronger impact on IBs performance during the pandemic. One potential explanation for this result is that Islamic banking institutions entered the “global financial crisis” in 2020 with better efficiency and higher capitalization levels (Mirzaei et al. 2024). Another explanation is that “unlike conventional securitization, which is marked by significantly low bank stability, an issuance of Islamic securitization leads to lower bank risk” (Abdelsalam et al. 2022, p.1). Therefore, a further increase in market concentration and bank efficiency is not necessarily associated with improved profitability of these financial institutions. The moderating effect of market concentration is negative and of marginal significance, supporting the notion that “banks with better efficiency are more profitable [..] in more competitive environments” (Mateev et al. 2023, p. 23). Therefore, in such circumstances, efficient banks are likely to be better positioned to mitigate the negative impact of the SAR-COV-2 pandemic, as markets with higher level of competitiveness create favorable conditions for these banks to improve their performance during the crisis.

The effects of market concentration and efficiency on bank stability

Our study also examines the impact of market concentration and efficiency on bank stability, and whether these effects are more pronounced during the SAR-COV-2 pandemic. To analyze bank stability and soundness, we used the following measures: distance to default (Z-score), multilevel performance index (MLPS), and financial stability indicator (FSI). Tables 5 and 6 illustrate the results for various banking systems (IBs and CBs) during the pandemic and the pandemic periods.

Table 5 Panel regressions of bank stability (CB vs IB, 2006–2019).

Table 6 Panel regressions of bank stability (CB vs IB, 2020–2021).

Table 5 shows that the coefficient estimates for concentration and efficiency variables are positive and possess statistical significance. This result suggests that increased levels of efficiency and concentration are associated with enhanced financial stability of MENA banks, as indicated by both the default risk (Z-score) and MLPS indicator. However, the expected signs of concentration and efficiency effects are opposite to our predictions when the dependent variable is the short-term financial stability (FSI). To better understand the reasons for these results, we calculated FSI for each country and found that FSI values were not homogeneous across the two samples and negative in the case of IBs, which indicates a substantial credit risk related to NPLs, particularly during the pandemic. Moreover, the results for FSI reported in Models 3 and 6 are statistically insignificant. One possible explanation is the fact that IBs faced significant credit risk associated with non-performing loan portfolios at the onset of the pandemic. This was exacerbated by the fact that Islamic institutions in some countries in the MENA region lacked access to certain liquidity management measures announced by central banks, “either due to the absence of credit lines for Islamic banks or legal impediments” (Rizwan et al. 2022, p. 2). In overall, our results for banks in the MENA region align with the findings for other developing economies. For instance, Miklaszewska et al. (2021) in their study on the CENE region, discovered an insignificant relationship between market concentration (proxied by the HH-index) and probability of default (Z-score). Similarly to our study they found that the link between market concentration and the aggregate performance indicator (MLPS) was marginally significant and positive, but negative for the FSI indicator.

The larger estimated coefficients of our variables of main interest indicate that the reported associations with bank stability are stronger for IBs than for CBs, which aligns with our expectations. The improved efficiency and stability of IBs in the “pre-pandemic period” can be attributed to changes in regulations and market conditions in the MENA countries. These factors compelled IBs to adopt best practices and enhance operational efficiency to maintain competitiveness (Otero et al. 2020). As a result, their ability to withstand adverse shocks such as the SAR-COV-2 pandemic has increased.⁸ The statistical significance of the moderating effect of market concentration on efficiency is negative for both types of banks. This suggests that efficient banks are able to achieve better financial stability when operating in less concentrated markets; notably, this observation remains consistent regardless of their business model. Our findings align with the predictions of the “competition-stability theory” (Boyd and De Nicolo, 2005) which suggests that competition decreases the volume of “non-performing loans” while increasing the stability of the bank. This effect is particularly noticeable for banks that demonstrate higher levels of efficiency. However, our results contradict those of González et al. (2017), who confirmed “the importance of the market structure as an explanatory factor for the financial stability of banks in the MENA region, but also show that higher concentrations do not have to be associated with less competitive markets as predicted by the SCP hypothesis” (p. 10).

Table 6 reports the results for bank stability during the SAR-COV-2 pandemic. We observed a notable difference in how market concentration and efficiency influence the stability of IBs and CBs during the pandemic. The significantly positive effect on IBs performance (measured by the Z-score and the two aggregate performance indicators) provides strong support to our second hypothesis (H2b). This finding aligns with the competition-fragility nexus which can be explained by the competition hypothesis. Two arguments support the view that high competition may increase banks fragility (Allen and Gale, 2004; Hellmann et al. 2000; Keeley, 1990). The first argument is that “high competition in the financial market erodes market power, lowers the profit margin and capital buffer, and results in reduced franchise value” (Md. Noman et al, 2017, p. 3). In response, banks might be motivated to assume more risk to offset declines in their franchise value and profits (Marcus, 1984; Keeley, 1990). The second argument suggests that in competitive markets, diminished financial rewards from intermediation could lead banks to reduce their scrutiny of potential borrowers, possibly lowering loan quality and compromising the bank’s stability. Our findings indicate that if concentrated markets are associated with a lower degree of competition, increased concentration will compel banks to enhance their financial stability.

The validity of our results is confirmed by the fact that both market concentration and efficiency effects are not only statistically significant but also economically relevant for banking stability. Data in Table 5 indicates that a “one standard deviation increase” in market concentration (13%), causes an increase of 9.035% (=0.695*13%) of bank stability for CBs and 97.27% (=7.482*13%) for IBs (see Models 1 and 4). Likewise, the concentration effects are economically relevant for both types of banks (CBs and IBs) when the overall performance is measured by the MLPS index (see Models 2 and 5). However, in the context of banks' short-term stability measured by the FSI index, the economic relevance of the concentration effect is insignificant, as evident from the estimated coefficients of the HHI measure (see Models 3 and 6). The analysis of bank stability during the crisis period (2020–2021) indicates the concentration effect is economically significant only for IBs (see Table 6). For instance, a “one standard deviation increase” in market concentration (13%) results in a 5.24% increase (=0.403*13%) in banks’ short-term stability (see Model 6). Similarly, efficiency also shows a significant economic effect.

When examining bank performance in the MENA region, certain findings of our analysis are notably significant. First, the analysis has highlighted that concentration and efficiency effects are particularly pronounced within the sample of IBs. Our explanation is that IBs possess a unique business model “in terms of their asset-liability structure and product offering.” Second, MENA countries with dominant Islamic banking markets have seen a rise in concentration levels as a result of the SAR-COV-2 pandemic, much like other developing regions (Hamadi and Awdeh, 2020). These two factors have enabled IBs operating in concentrated markets to demonstrate increased financial stability compared to their conventional peers. Third, since enhanced efficiency (and concentration) leads to better financial stability, new governance-related initiatives grounded on Sharia’s values should be implemented as factors “enabling Islamic banking institutions to mitigate the negative effects of global financial crises such as the COVID-19 pandemic” (Abdulla and Ebrahim, 2022, p. 240).

The effects of market concentration and efficiency on bank capitalization

Tables 7 and 8 display the results of the regression analysis for banks’ capital positions (see model specification outlined in Eq. 2). We use three different measures of a bank’s capitalization level, specifically “equity to total assets (E/TA), capital adequacy ratio (CAR), and total capital requirements (TCR)”.

Table 7 Panel regressions of bank capitalization (CB vs IB, 2006–2019).

Table 8 Panel regressions of bank capitalization (CB vs IB, 2020–2021).

Our analysis yields several interesting results. Firstly, it is evident that prior to the SAR-COV-2 pandemic, both concentration and efficiency had a notable impact on the capital position of MENA banks, which confirms our last hypothesis (H3a). However, these effects are statistically significant only in the sample of CBs (see Models 1 and 2). Secondly, during the pandemic, concentration and efficiency had a significantly positive effect on “capital adequacy ratio” (CAR) but a negative impact on “equity to total assets ratio” (E/TA). According to prior research, these effects can be credited to the recent financial crisis, “which resulted in a decrease in equity and an increase in regulatory capital” (Miklaszewska et al. 2021, p.16) as a response to the need to establish “a sufficient capital conservation buffer to mitigate the impact of downside risk during the pandemic” (Demirgüç-Kunt et al. 2021). In line with our predictions, these effects are more pronounced within the sample of IBs, which aligns with our last (H3b) hypothesis. Market concentration shows a significant moderating effect which is negative during the pandemic period, indicating that efficient banks operating in more competitive markets are also better capitalized, particularly when the banking institution is Islamic.

Our results revealed that during the SAR-COV-2 pandemic, bank characteristics such as loans as a percentage of deposits, loan size, liquid assets, and “non-performing loans” had a significantly positive effect on banks' capital ratios. Conversely, non-financial loan growth and non-interest income were negatively correlated to the capital adequacy level of the MENA banks. Thus, our empirical results failed to sustain the idea that “traditional bank business models based on intermediation do not guarantee satisfactory profitability [..], but also deplete banks’ capital basis. This observation is supported by the positive impact of non-interest incomes on bank capital ratios” (Miklaszewska et al. 2021, p.16). On the opposite, our findings imply that MENA banks do not rely heavily on “non-interest revenue sources as part of their income diversification strategy” during the SAR-COV-2 pandemic. The analysis outcomes have strong policy implications for regulatory authority which should mandate bank managers to adopt a more disciplined approach towards lending decisions and request to establish a “sufficient capital conservation buffer.” This is remarkably crucial in the context of the SAR-COV-2 pandemic when banks face serious liquidity issues.

Alternative tests and robustness checks

We conducted several robustness tests to verify the validity of our results. Because the econometric models employed in this study are “dynamic panel models”, using conventional econometric techniques like OLS and FE estimators can be challenging due to the “possible correlation between the delayed endogenous variable and the unobservable effects” (Neto, 2021, p. 114), which can result in biased and inconsistent estimates. Arellano and Bond (1991) and Blundell and Bond (1998) proposed the “General Method of Moments (GMM)” estimator as a solution to this problem. In the empirical literature, the GMM estimators can be categorized as “difference GMM estimator” and “system GMM estimator” developed by Arellano and Bover (1995) and Blundell and Bond (1998), respectively (Neto, 2021). García-Herrero et al. (2009) reported that “the GMM estimator system is capable of accounting for factors such as potential endogeneity, unobserved heterogeneity, as well as the persistence of the dependent variable” (p.10). Therefore, in this study, we employed one-step “GMM estimator system” to estimate our reference model (see Eq. (1) and Eq. (2)). We also conducted a Sargan-Hansen test to evaluate “the accuracy of our instruments” and to ensure that over-identifying restrictions were met. Finally, we apply Arellano and Bond’s test to identify the non-existence of second-order autocorrelation, AR(2). Tables 9 and 10 display the results of this analysis for the pre-pandemic and pandemic periods, respectively. Our earlier findings for the pre-pandemic period, reported in Table 3, are confirmed, demonstrating a significantly positive correlation between market concentration, efficiency, and bank profitability. During the pandemic these effects are significant only in the sample of CBs, which coincide with our results reported in Table 4. The moderating effect of concentration is marginally significant and negative for both pre-pandemic and pandemic periods. Our findings indicate a strong positive effect of efficiency on bank performance for either type of bank.

Table 9 Panel regressions (one-step GMM estimator) of bank profitability (CB vs IB, 2006–2019).

Table 10 Panel regressions (GMM estimator) of bank profitability (CB vs IB, 2020–2021).

Next, our robustness tests involved a variety of measures for our independent variables. First, we employ the “five-bank concentration ratio” (CR5) as a substitute for banking sector concentration. Our first hypothesis posited that market concentration and efficiency should exert a significant and positive influence on profitability. Moreover, we anticipated this effect to be stronger for Islamic banking institutions and during the SAR-COV-2 pandemic. The results reported in Tables S5 and S6 (see Supplementary Materials) confirmed these expectations only for the pre-crisis period. Second, we use the “net interest margin (NIM)” ratio as another proxy for bank profitability and growth. The analysis using the NIM variable supports Tan et al. (2017) who reported that “efficiency plays a significant and negative impact on bank profitability when measured by ROA, but a positive impact on ROE and NIM” (p.12). Following previous research (see e.g., Abdallah and Ismail, 2017), we use Tobin’s Q as “a market-based proxy” for bank performance. The results (not reported here due to space limitation) confirmed that concentration and efficiency effects are more pronounced during the SAR-COV-2 pandemic, particularly in the sample of IBs.

Finally, instead of using two separate periods (before and during SAR-COV-2), we included a crisis dummy variable “which takes value of 1 for the years 2020 and 2021, and 0 otherwise”. Regression outcomes reported in Table S7, (see Supplementary Materials) indicate a strong negative impact of the pandemic on banks profitability in the MENA region. Moreover, we find a significant interaction effect of the pandemic variable with concentration and efficiency on bank profitability (the coefficient estimates of the two interaction terms with “the crisis dummy” are significantly negative). This suggests that the positive impact of concentration and efficiency on bank performance deteriorates during the SAR-COV-2 pandemic. Although the pandemic shock significantly reduced banks’ profitability, it was not detrimental to MENA banks’ stability (see Table S2).

We also investigate whether market concentration and financial performance are “endogenously determined” using the “two-stage least squares (2SLS)” approach (Farag et al. 2018). Table S8 (see Supplementary Materials) presents the estimation results of the “instrumental variables (IV)” regressions conducted using the 2SLS method. As expected, the instrument (three lags of the HH-index) shows a high level of significance in the “first-stage” regression. In the “second stage”, we observe a robust and positive correlation between the “fitted value of the HH-index” and the “financial performance, as measured by ROA and ROE”, across different samples (i.e., Islamic and conventional banks). The results align with those reported in Tables 3 and 4, indicating that banks operating in concentrated markets with less competition achieve better financial performance. This effect is particularly notable among IBs. To address potential endogeneity issues related to the relationship between both concentration and efficiency and financial performance, we conducted a Hausman test for each model in Table S8 that includes our profitability measures (ROA, ROE, and C/I) and reported the endogeneity estimates. To converse space, we reported only the difference between the coefficients of both regressions, that is, the endogeneity bias estimates from the Hausman test outputs, along with the related statistics (Table S9). The results for the Hausman test fail to reject the “null hypothesis” that efficiency is exogenous. Moreover, our instrument passed the “Stock and Yogo (2005)” test for weak instruments, indicating its validity and reliability in the instrumental variables (IV) analysis.

Discussions and conclusions

Numerous studies have examined the influence of market concentration and efficiency on the financial stability of banks. However, no econometric study has been undertaken to examine how market concentration and efficiency, both individually and jointly, affect the performance of banking sector in the MENA region. Our study is pioneering in distinguishing between the Islamic and conventional banking systems co-existing in the MENA region. We have discovered that concentrated banks and those with improved efficiency demonstrate better financial stability during the SAR-COV-2 pandemic.

Using a comprehensive dataset of 575 banks spanning 20 countries in the MENA region, this research delves into the effects of market concentration and efficiency on the financial performance and stability of MENA banks. Our findings indicate that, prior to the SAR-COV-2 pandemic, both concentration and efficiency positively influenced the financial performance of banks in the MENA region. Specifically, banks that were more efficient and operated in concentrated markets displayed better profitability and improved cost-effectiveness. During the SAR-COV-2 pandemic, a substantial differential impact of concentration and efficiency on IBs performance is observed. The enhanced performance and stability of IBs during the pandemic can be attributed to the changes in regulations and market conditions in the MENA countries. Moreover, the ‘asset-backed business models” and stronger capitalization have empowered IBs to better withstand the adverse effects of the SAR-COV-2 pandemic.

Our study significantly contributes to the current research landscape by exploring the efficiency-default risk paradigm within the context of Islamic banking. While previous studies have shown that improved efficiency can positively impact financial stability and decrease credit risk for banks (Isnurhadi et al. 2021; Mateev et al. 2022b), we extend their scope by adopting a unique approach that considers both individual and aggregate measures of bank stability and soundness. Our research findings provide strong evidence that, prior to the pandemic, market concentration and efficiency were significant “determinants” of the financial stability of banks in the MENA region. However, during the SAR-COV-2 period, these positive effects were observed exclusively in the sample of IBs. Additionally, we found that less concentrated markets, typically associated with increased market competition, positively moderate the association between bank efficiency and financial stability during the pandemic. These findings have their evidence rooted in the predictions of the “competition-stability” theory and the results of previous empirical studies reporting that competition increases bank stability as it decreases the volume of non-performing loans (Mirzaei et al. 2013; González et al. 2017, Mateev et al. 2022b).

During the SAR-COV-2 pandemic, both market concentration and efficiency significantly impacted the capital adequacy of banks in the MENA region. Our analysis of different banking systems indicates that, prior to the pandemic, concentration and efficiency did not contribute to improving the capital position of IBs. However, the further consolidation in the Islamic banking market caused by the pandemic has had a positive impact on the capitalization level of these banks. Moreover, the negative impact of non-interest income on capital ratios indicates that banks in the MENA region do not heavily rely on non-interest revenue sources as part of their income diversification strategy during the pandemic. Therefore, our findings do call for adjustment of the current banking models that heavily rely on intermediation and interest-based earnings, especially for CBs. This alteration is not as urgent for Islamic banking institutions in the MENA region, as they entered the global financial crisis in 2020 with better efficiency and higher levels of capitalization compared to their conventional peers (Mirzaei et al. 2024).

Policy implications

Bank managers, policymakers, and regulators in the MENA region should pay attention to the implications of our research. First, bank managers should focus on developing strategies that promote bank efficiency since financial performance is positively associated with the efficiency level of banks. Second, policymakers and regulatory authorities need to establish the necessary support policies to assist Islamic banks in recovering in the post-SAR-COV-2 pandemic. For example, further consolidation in this market may positively influence bank stability and soundness. However, the regulatory institutions should be cautious about implementing such policies, considering their potential adverse impacts on competition, bank customers, and overall stability. Third, decision-makers should require managers to take a more disciplined approach to lending decisions and mandate the establishment of a sufficient capital conservation buffer to mitigate the impact of liquidity risk during the pandemic. In overall, our evaluations recommend a possible alteration of current banking models that rely on intermediation and interest-based earnings, particularly for CBs. Moreover, it is imperative for IBs to consider the use of non-interest-bearing revenue streams as a significant component of their risk diversification plans in the post-SAR-COV-2 period. These recommendations may serve as valuable guidelines for policymakers and governing bodies in other nations where Islamic and conventional bank services are offered hand-in-hand during the SAR-COV-2 pandemic.

Limitations and call for future research

Our analysis is constrained by a key limitation related to the fact that the influence of the recent financial crisis was studied over a restricted time frame of two years. Therefore, further investigation of the SAR-COV-2 effects on bank performance will require an extended period of analysis including the year 2022. This study provides recommendations for future research in this field, proposing several potential avenues for discovery. First, there is a need to explore the influence of additional moderating factors beyond concentration. Future research may delve into the impact of other variables, such as ownership concentration, market size, and liquidity as these may also significantly influence the relationship between efficiency and bank performance. Second, to achieve a more comprehensive understanding of this relationship from a global perspective, future studies may consider conducting similar research on different regions with the prevailing presence of Islamic banking.